Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Computing | en_US |
| dc.contributor.advisor | Chan, Keith (COMP) | - |
| dc.creator | Sin, Sau Mei | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/10158 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | - |
| dc.rights | All rights reserved | en_US |
| dc.title | Deep learning for morellian analysis for art connoisseurship | en_US |
| dcterms.abstract | Art sales is really a promising business. Art collectors and museums are willing to pay billion dollars to acquire a piece of canvas if it is proven as a historically significant masterpiece. Bringing fame and fortune, it is no doubt that forgery is a very attractive business. It is not just a serious crime, but could be an embarrassing scandal to a nation, if a world class museum waste a huge amount of public money in acquiring a fake. As a crucial role to detect the forgery, art connoisseurship has been relying on art experts and human eyes throughout the art history. Art historians and curators spent their whole life in art literatures in order to cultivate themselves and train their eyes to be very sensitive to particular artist and artistic style. Scientific methodology such as x-ray technology, UV light, microscope, infrared reflectography, were developed to assist art experts to see beyond the canvas. However, forgery is still threatening the art world today. With technology advances, computer scientists tried hard to develop computer vision model and apply it to art connoisseurship. Yet, the recent researches have neglected the importance of domain knowledge. Art history, visual culture and aesthetics, such theory provide the fundamental to human vision, perception and interpretation of visual art. Interdisciplinary studies across computer vision and visual studies is essential in designing the models to conduct automated art connoisseurship. Filling the gap, this project is going to conduct an interdisciplinary research to bridge the human vision and computer vision. From the aspects of art history and computer vision, the project will provide a new perspective of effectively utilising computer vision technology in analysing fine art. The project will imitate the Morellian Analysis, which is commonly used by art historians to analyse stylistic features of artists, by a multi label CNN classification model. It is the real practical solution, facilitating art research and art business by the-state-of-the art computer vision technology. | en_US |
| dcterms.extent | x, 99 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2019 | en_US |
| dcterms.educationalLevel | M.Sc. | en_US |
| dcterms.educationalLevel | All Master | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.LCSH | Art -- Forgeries | en_US |
| dcterms.LCSH | Art and technology | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 991022268443403411.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 4.25 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/10158